CANDU Steam Generator Life Management

A> AECL EACL

AECL-11912 CA0000120

R.L. Tapping, J. Nickerson, P. Spekkens, C. Maruska

1998 January AECL

CANDU STEAM GENERATOR LIFE MANAGEMENT

R.L. Tapping*, J. Nickerson* P. Spekkens+and C. Maruska+

AECL Chalk River Laboratories Chalk River, Ontario, CANADA, KOJ 1J0

*AECL Sheridan Park Research Community 2251 Speakman Drive Mississauga, Ontario, CANADA, L5K 1B2

tOntario Hydro Chemistry and Metallurgy Department 700 University Avenue Toronto, Ontario, CANADA, M5G 1X6

1998 January

AECL-11912 EACL

GESTION DE LA DURÉE DE VIE DES GÉNÉRATEURS DE VAPEUR CANDU

par

R.L. Tapping*, J. Nickerson**, P. Spekkens+ et C. Maruska+

RÉSUMÉ

Les générateurs de vapeur constituent un élément crucial d'un réacteur de puissance et peuvent contribuer considérablement à l'indisponibilité d'une centrale, comme on l'a amplement démontré dans les réacteurs à eau sous pression (PWR). Les générateurs de vapeur CANDU ne sont pas à l'abri de détériorations et la diversité des modèles de générateurs de vapeur CANDUMD et des matériaux des tubes ont présenté quelques problèmes inattendus. Toutefois, des mesures correctives énergiques et des travaux d'entretien proactifs vigilants ont eu pour résultat une réduction de l'indisponibilité des centrales canadiennes CANDU en raison des générateurs de vapeur.

EACL et les compagnies d'électricité exploitant des centrales CANDU ont défini des programmes qui permettront aux nouveaux générateurs de vapeur ou à ceux existants de fonctionner efficacement pendant 40 ans. Les travaux de recherche et développement portent sur la corrosion et la détérioration mécanique des faisceaux de tubes et des pièces internes, la chimie, la thermohydraulique, l'encrassement, l'inspection et le nettoyage ainsi que sur la mise au point d'outils spéciaux destinés à résoudre des problèmes ponctuels. Un élément moteur important est la mise au point de directives d'aptitude au service propres à la filière CANDU, et notamment les techniques de surveillance et d'inspection appropriées pour contrôler l'état des générateurs de vapeur. Le présent rapport montre également de quelle façon les progrès tout récents en matière de technique de nettoyage font partie de la stratégie de gestion pendant la durée de vie. Les travaux à plus long terme sont axés sur la mise au point de la surveillance en ligne intelligente du circuit d'eau d'alimentation et des générateurs de vapeur. Les nouveaux modèles de générateurs de vapeur présentent des risques moindres de corrosion et d'encrassement, de plus grandes facilités d'inspection et de nettoyage et sont moins susceptibles aux dommages mécaniques.

Les compagnies d'électricité canadiennes qui exploitent des centrales CANDU ont élaboré des programmes de mesures correctives pour lutter contre les baisses de rendement (Gentilly 2, Pointe Lepreau, Bruce A et B, Pickering A et B) et ont élaboré des programmes stratégiques pour garantir à l'avenir une bonne exploitation. Le programme de recherche et développement et l'expérience pratique acquise en exploitation ont permis de déterminer les domaines dans lesquels on peut apporter des améliorations aux méthodes d'exploitation ou de conception afin d'assurer l'exploitation des générateurs de vapeur pendant toute leur vie nominale à un facteur de capacité acceptable.

*EACL Laboratoires de Chalk River Chalk River (Ontario) CANADA KOJ 1J0

"EACL Sheridan Park Research Community 2251, rue Speakman Mississauga (Ontario) CANADA L5K 1B2

+Ontario Hydro Chimie et Métallurgie 700, avenue University

Toronto (Ontario) CANADA M5G 1X6

1998 Janvier

Md Réacteurs Canada Deuterium Uranium AECL-11912 AECL

CANDU STEAM GENERATOR LIFE MANAGEMENT

R.L. Tapping*, J. Nickerson" P. Spekkens+and C. Maruska*

ABSTRACT

Steam generators are a critical component of a nuclear power reactor, and can contribute significantly to station unavailability, as has been amply demonstrated in Pressurized Water Reactors (PWRs). CANDU steam generators are not immune to steam generator degradation, and the variety of CANDU® steam generator designs and tube materials has led to some unexpected challenges. However, aggressive remedial actions, and careful proactive maintenance activities, have resulted in a decrease in steam generator-related station unavailability of Canadian CANDU reactors.

AECL and the CANDU utilities have defined programs that will enable existing or new steam generators to operate effectively for 40 years. Research and development work covers corrosion and mechanical degradation of tube bundles and internals, chemistry, thermalhydraulics, fouling, inspection and cleaning, as well as provision for specialty tool development for specific problem-solving. A major driving force is development of CANDU-specific fitness-for-service guidelines, including appropriate inspection and monitoring technology to measure steam generator condition. This paper will also show how recent advances in cleaning technology are integrated into a life management strategy. Longer-range work focuses on development of intelligent on-line monitoring for the feedwater system and steam generator. New steam generator designs have reduced risk of corrosion and fouling, are more easily inspected and cleaned, and are less susceptible to mechanical damage.

The Canadian CANDU utilities have developed programs for remedial actions to combat degradation of performance (Gentilly-2, Point Lepreau, Bruce A/B, Pickering A/B) and have developed strategic plans to ensure that good future operation is ensured. The research and development program, as well as operating experience, has identified where improvements in operating practices and/or designs can be made in order to ensure steam generator design life at an acceptable capacity factor.

*AECL Chalk River Laboratories Chalk River, Ontario, CANADA, KOJ 1J0

"AECL Sheridan Park Research Community 2251 Speakman Drive Mississauga, Ontario, CANADA, L5K 1B2

+Ontario Hydro Chemistry and Metallurgy Department 700 University Avenue Toronto, Ontario, CANADA. M5G 1X6

1998 January

Canada Deuterium Uranium reactors AECL-11912

TABLE OF CONTENTS Page

1. INTRODUCTION 1

2. CANDU UTILITY EXPERIENCE 1

3. CANADIAN CANDU LIFE MANAGEMENT STRATEGIES 2

4. RESEARCH AND DEVELOPMENT ACTIVITIES 4

5. SUMMARY 6

6. ACKNOWLEDGMENTS 6

7. REFERENCES 6

LIST OF TABLES

Table 1: Summary of design and chemistry specifications for Canadian CANDU stations 8

Table 2: Summary of degradation mechanisms found in CANDU steam generators to date 8

Table 3: Summary of life management strategy highlights for Canadian CANDU steam generators 10

Table 4: Summary of R&D activities in the COG steam generator program 11

LIST OF FIGURES

Figure 1: Ontario Hydro Nuclear station incapability due to steam generator forced and planned outages 12

Figure 2: Key elements of a steam generator life management program and associated interfaces 13 1. INTRODUCTION

Until recently CANDU steam generators had an enviable record: virtually no tubes plugged because of corrosion [1]. Since then several Ontario Hydro reactors have experienced steam generator corrosion, fouling degradation, and fretting of the upper bundle. This has led to thermal and hydraulic inefficiencies as well as tube integrity concerns. To date most CANDU-6 (C-6) steam generators have not experienced significant corrosion degradation, although reactor inlet header temperature (RIHT) increases have required remedial action at both Gentilly-2 and Point Lepreau Nuclear Generating Station (NGS), and Point Lepreau has had some minor steam generator tube corrosion.

Hence CANDU steam generators are subject to many of the same problems that have plagued pressurized water reactor (PWRs) for the past 20 years, although to a much lesser degree. This has led to a concerted effort to restore the health of those steam generators that have been degraded, and to proactive measures to ensure that all CANDU steam generators, current and future, meet their design lifetimes. Supporting these efforts is a research and development (R&D) program aimed at providing the knowledge necessary to understand, predict and prevent steam generator degradation.

CANDU steam generators are smaller and different in design than those used in PWRs. Also there are differences in materials, and primary- and secondary-side chemistries, and CANDU steam generators operate at lower temperatures than those in PWRs. AECL and the CANDU utilities have had to develop their own solutions to steam generator degradation to ensure an effective steam generator service life. Table 1 summarizes the Canadian CANDU design and chemistry features.

This report briefly describes the CANDU utility experience with steam generators, the various CANDU steam generator life management strategies that have been put into place, and the supporting R&D program. The report also provides design recommendations for new steam generators.

2. CANDU UTILITY EXPERIENCE

Until the mid 1980's fouling of the secondary side of long service Ontario Hydro Nuclear (OHN) CANDU steam generators was significant, but with little apparent loss of function. By 1990 the support plate flow holes at Bruce-A NGS were -80% blocked [2,3]. This led to flow instabilities (boiler level oscillations) and a program of secondary side cleaning was initiated. Soon after this, under-deposit corrosion of steam generator tubes in the sludge pile region was found at Pickering-B NGS. At approximately the same time the effects of lead contamination on the Bruce Unit 2 steam generator became evident. Later, sludge pile pitting was discovered at Pickering 1, and fretting became a concern at Bruce-B NGS. These experiences are summarized in Table 2, along with CANDU 6 data. They indicate that prolonged neglect of fouling of CANDU steam generators will lead to thermal performance and corrosion degradation. Repeated detection of fretting wear also indicates that the fabrication of some CANDU steam generators led to some susceptibility to fretting damage in the U-bend.

In addition to tube degradation, a concern with degradation of primary side divider plates, leading to excessive leakage from inlet to outlet, has emerged recently. Divider plates have been replaced recently at Point Lepreau and Gentilly-2 in 1994/95, and this has had a significant positive impact on steam generator thermal performance.

Despite these problems, the CANDU utilities were able to restore steam generator performance in most units using a program of extensive cleaning, often accompanied with refurbishment. Figure 1 illustrates the improvement since 1992, in terms of steam generator contributions to incapacity factor, for OHN -2- stations. The intent of the refurbishment was to remove copper alloys from the feedwater train, and to reduce the risk of impurity ingress. The remedial actions are also summarized in Table 3. (It is now anticipated that, with the exception of Bruce Unit 2, and with an appropriate life management strategy in place, steam generators should not become the station-life-limiting component in CANDU stations.) At Bruce Unit 2 the extent of lead-induced stress corrosion cracking, and lead contamination, was such that the unit was shut down in 1995 September. Low demand for electricity in Ontario and the projected uncertainty in tube life led to a decision to cease operation. It is anticipated that, should electricity demand, and economics, permit, the Unit 2 steam generators and fuel channels will be replaced and the unit returned to power.

Concerns about U-bend vibration and stress, which have led to fretting, fatigue and stress corrosion cracking of tubes in the U-bend of various steam generators, have led to improvements in U-bend support structures and designs. At Bruce-A, anti-vibration bars (AVBs) were added to stabilize the structure to reduce the risk of mechanical fatigue, stress corrosion cracking and corrosion fatigue. Similarly, at Bruce-B, the installation of AVBs was required to reduce fretting degradation. More recent designs (Darlington NGS (DNGS), Wolsong-2, -3, -4) use stainless steel flat bar U-bend supports, to provide a greater margin against support-induced stresses on the U-bend tubes than do scallop bar supports. However, the effect on fretting wear, compared to correctly installed scallop bar supports, has not yet been determined.

3. CANADIAN CANDU LIFE MANAGEMENT STRATEGIES

All Canadian CANDU utilities have a steam generator life management strategy. Those utilities that have experienced degradation (Bruce A, Pickering A and B, Point Lepreau, Gentilly-2, Bruce-B) have already carried out remedial actions, as indicated in Table 3, and DNGS has initiated a proactive maintenance strategy. Many of these stations have developed a formal strategic planning document which has been peer reviewed, and the others have operational and maintenance plans which address the same need. Table 3 summarizes the key components of each station's steam generator life management strategy. All stations address the current Electric Power Research Institute (EPRI) recommendations [4] where they are relevant for each individual station. The major goals of the plans are to:

• avoid surprises, • learn from experience, and • manage the investment in steam generators.

Figure 2 summarizes the key components needed to implement a life management strategy.

The methodology necessary to put a life management strategy in place is first to establish the station objectives with respect to expected service life and required reliability. It is then important to characterize the plausible degradation mechanisms that may affect steam generator life and reliability, and how these may affect station life and reliability. Important here is a sound fitness-for-service (FFS) plan, and good inspection capability. Steam generator life and reliability is then a function of available remedial actions and a preventive maintenance strategy. This may require equipment modifications, cleaning strategies, effective monitoring programs, and improved chemistry control. Clearly cost effectiveness is important, and may vary with time depending on station economics and power demand. Each of the plans summarized in Table 3 has these elements, with the ratio of remedial action to proactive maintenance varying from station to station. Currently Bruce A is still in the remedial action mode, whereas Gentilly-2 and Pickering B are transitioning to the proactive maintenance stage, and Darlington is in the proactive maintenance stage. -3-

Remedial programs have paid particular attention to chemistry control, with upgrading of water treatment plants and condenser upgrades (or upgraded operating procedures such that condenser leaks are no longer tolerated) a major consideration. Chemistry specifications have been tightened, and Ontario Hydro stations place heavy emphasis on maintaining secondary chemistry specifications within guidelines.

The steam generator remedial actions have largely been based on extensive cleaning campaigns, including the use of specialized water lance equipment and low- and high-temperature chemical cleaning, and crevice cleaning methods. All of these had to be specifically tailored to the CANDU designs, materials and sludge characteristics. CANDU specific cleaning developments are described in another paper at this conference.

For CANDU reactors to benefit fully from their on-power refueling capability, steam generator maintenance must be carried out during the annual, or in some cases biannual, maintenance outages. There may be a need to have an extended maintenance outage for chemical cleaning every 10 years or so, depending on the fouling situation. Many CANDU stations are currently carrying out remedial and rehabilitative actions to redress years of accumulation of deposits. It has been noted that the cost of these activities is similar to the cost of putting this into place, after degradation has occurred, and maintaining effective long-term maintenance strategies designed to avoid future serious degradation [5]. The Darlington steam generator life assurance plan notes that the cost of an extensive proactive maintenance strategy designed to avoid serious degradation will be about 10% of the cost, over the 40-year station life, of rehabilitating the steam generators after they have been subjected to chronic fouling and impurity ingress. Thus it is evident that routine, frequent and, as necessary, enhanced inspection is an essential part of any effective maintenance strategy, along with a good proactive plugging policy. The past measure that no steam generator leaks implied good performance, with little or no knowledge of the actual condition of the steam generator, is now unacceptable and costly.

Coupled with inspection, is the need for a sound and flexible FFS program that addresses plausible station degradation mechanisms. For current CANDU steam generators the variability of designs implies that fouling and corrosion degradation mechanisms will vary from station to station and hence each station will require a different FFS guideline as well as station specific life management plans.

An important part of the OHN FFS is tube removal to validate degradation mechanisms and assist with root cause analysis. In one case, at Bruce Unit 2, sufficient tubes (-100) were removed that the metallographic analyses could be used to validate the inspection results. This provides considerable confidence, both for the utility and the regulator, in the FFS and life management strategies.

From a design perspective there are several factors to consider when striving for a 40-year steam generator life. Perhaps the most important factor, after good chemistry control (and specifications and equipment to assure this), is the need for a leak-tight condenser. For seawater-cooled sites a titanium- tubed condenser with leak-tight tube-to-tubesheet joints (such as Wolsong-1) is recommended to minimize any risk of impurity in-leakage, although few utilities have these. It is often assumed that for fresh-water sites the condenser requirements can be relaxed, but experience at CANDUs and PWRs indicates that this assumption is false [6]. Although the risk of chloride ingress from lakewater in- leakage may be lower, the concentration over time of chloride in steam generator crevices will eventually reach the same levels as seawater in-leakage. Thus leak-tight condensers for fresh water-cooled sites should be as high a priority as for seawater-cooled sites. Other issues related to improved designs for long-term steam generator reliability are briefly outlined in Section 4. -4- -5-

4. RESEARCH AND DEVELOPMENT ACTIVITIES

Within the CANDU Owners Group (COG) there is an R&D program aimed at improving current steam generator reliability and operability, and some of this information is used to improve future steam generator designs, which would include replacement steam generators if applicable. AECL has a smaller program focussed on design improvements. The entire program covers all aspects of steam generator and balance-of-plant technology, with the primary focus on corrosion, fouling, cleaning, vibration, specialized tooling and non-destructive inspection. Table 4 summarizes the important activities currently underway. Major goals are to put in place the data and tools needed for effective FFS guidelines, to provide effective support for CANDU secondary system chemistry specifications, and to provide input into design requirements, in particular the data needed to ensure that steam generator design life is economically achievable. There is considerable effort aimed at providing steam generator operators with the rationale and tools to improve steam generator performance, especially corrosion, fouling and fretting mechanisms and inspection tooling. This knowledge, combined with that learned from steam generator tube removals, also provides FFS arguments, an important contribution to life management activities. In what follows, some recent R&D developments that contribute to design and life management strategies are outlined.

Downcomer flow is a measure of steam generator health. In a recirculating steam generator it indicates steam generator support structure fouling by providing an estimate of the recirculation ratio. Comparison with THIRST* code thermalhdyraulic predictions can then be made and an assessment of overall thermalhydraulic efficiencies derived. Downcomer flow measurements can be made in-situ during operation over the entire power range with a non-invasive ultrasonic technique developed by AECL. This technique also detects the presence of void in the downcomer and hence separator efficiency or other steam leak paths can be inferred. Downcomer flow measurements have been made at Bruce A and Darlington, and others are planned. These measurements suggest that cleaning of the Bruce A upper support plates has restored thermalhydraulic behaviour there, and that Darlington is operating as designed.

As noted earlier, chemical cleaning is now part of the CANDU steam generator life management strategy, both for existing and new stations. Research and development in cleaning technology has been oriented towards optimizing available low and high temperature processes for CANDU steam generators, and more recently towards developing dilute or mild cleaning for routine, low-cost and non-corrosive cleaning, probably during shutdowns. This work includes the development of cleaning technology for primary side steam generator fouling, which may be a contributor to increasing RIHT. This fouling also increases radiation fields in the vicinity of steam generators and negatively impacts non-destructive evaluation (NDE) inspection campaigns.

The NDE work has led to the development of eddy current probes capable of detecting axial or circumferential cracks in 0.5" and 0.625" tubes (C-5 and C-3 probes, respectively), of rotating probes for detection of pits in Monel 400 (Carter probes), and, more recently, of dual-purpose probes for rapid detection of cracks and volumetric defects. In parallel with this are developments of ultrasonic probes for more detailed analysis of degraded areas, especially those at the detection limit (or below) of eddy current.

There are also projects to improve data handling, since the C-3 and C-5 probes generate significant data that impacts on the ability of the analyst to complete the analysis quickly, and hence on overall

* Thermalhydraulics of Recirculating Steam Generators -6-

inspection time. If life management is to be successful, more rapid, and accurate, steam generator inspections are required.

Part of the NDE work, in collaboration with FFS, is validating NDE response and qualifying analysts; both activities are essential to an effective, and cost-efficient, life management program.

Hideout return studies are an important component of steam generator life management, and provide a direct measure of chemistry control. Such studies at a number of Ontario Hydro stations have shown that good chemistry control has reduced fouling and hideout of impurities [7]. Older units had considerably more impurities "hidden out" than newer ones, which is indicative of the greater degree of fouling in the older units. It also suggests that it is important to maintain a clean steam generator. CANDU stations typically do not operate under molar ratio control guidelines, but analysis of chemistry data for Ontario Hydro stations indicates an operating molar ratio of 0.6 to 0.8. At Bruce A, where boric acid additions are made to inhibit scallop bar corrosion, hideout return studies show that boric acid is indeed penetrating the crevices and deposits, as desired.

As Table 4 shows, there is considerably more R&D that is having an impact on steam generators, much of it aimed at improving inspection technology and quantifying design requirements and operating specifications. Changes in these requirements and specifications can be made only if there is a good understanding of the science and engineering behind them, so that their impacts can be properly assessed.

5. SUMMARY

While CANDU steam generators had an enviable record until 1988, recent experience has shown that they are not immune to some of the problems that have plagued PWR plants. To address these, the CANDU utilities have put in place aggressive steam generator life management programs. AECL and COG have undertaken proactive research and development work aimed at both supporting these life management programs and providing knowledge to lead designs for future units. It is recommended that a rigorous and detailed life management program be put in place at all CANDU stations. A detailed assessment of all potential degradation mechanisms is a key ingredient of the program.

6. ACKNOWLEDGMENTS

The authors acknowledge COG for its support of the steam generator program, and the stations for their assistance in dealing with steam generator issues over the years. Contributions of P.V. Balakrishnan (AECL) are gratefully acknowledged.

7. REFERENCES

1. R. Dyck, A. Marchand, P. Spekkens and K. Verma, "Operational Experience with Steam Generators in Canadian Nuclear Power Stations", Proc. Steam Generator and Heat Exchanger Conference, p. 1- 10 (Toronto, 1990, CNS).

2. J. Malaugh and S. Ryder, "Bruce NGS-A Support Plate Inspection and Waterlancing", Proc. Steam Generator and Heat Exchanger Conference, p. 3-51 (Toronto, 1990, CNS). -7-

3. S. Roy, "Steam Generator Performance and Life Management Decisions in Bruce A Nuclear Generator Station", Proc. Steam Generator and Heat Exchanger Conference, p. 1-47 (Toronto, 1994, CNS).

4. R.L. Tapping, "Steam Generator Life Management Strategy: An Overview", unpublished report, 1995 August.

5. K. Talbot, luncheon address to the CANDU Maintenance Conference, Toronto, 1995 November.

6. L.M. Stipan and R.L. Tapping, "Steam Generator Tube Performance: Experience with Water-Cooled Nuclear Power Reactors during 1988-1991", unpublished report, 1995 October.

7. P.V. Balakrishnan, S.M. Pagan, A.M. McKay and F. Gonzalez, "Hideout and Return: Laboratory Studies and Plant Measurements", paper presented at the Specialists' Meeting on Improving the Understanding and Control of Corrosion on Secondary Side of Steam Generators, Airlie, VA, 1995 October 9-13; Also CANDU Owners Group Report COG-I-95-555. -8-

Table 1: Summary of design and chemistry specifications for Canadian CANDU stations.

Plant History PICKERING A PICKERING B BRUCE A BRUCE B DARLINGTON CANDU 600 (4x542 MWe) (4x540 MWe) (4x904 MWe) (4x915 (4x935 MWe) POINT GENTILLY-2 MWe) LEPREAU (685 MWe) (680 MWe) Commercial Ul July'71 U5 May '83 Ul Sep'77 U5 Mar '85 Ul Nov'92 Feb '83 Oct '83 Operation U2 Dec '71 U6 Feb '84 U2 Sep '77 U6 Sep '84 U2 Oct '90 U3 June 72 U7 Jan '85 U3 Feb '78 U7 Apr '86 U3 Feb '93 U4 June '73 U8 Feb '86 U4 Jan '79 U8 May '87 U4 Jan '93

Number of tubes/SG 2600 2573 4200 4200 4663 3550 3550 Thermal Output/SG 138 MWt 138 MWt 280 MWt 280 MWt 664 MWt 516 MWt 516 MWt Tube Material Monel 400 Monel 400 Inconel 600 Inconel 600 Incoloy 800 Incoloy 800 Incoloy 800 Tube Size (OD) 12.7 mm 12.7 mm 12.7 mm 12.7 mm 15.9 mm 15.9 mm 15.9 mm TWTcold (°C) 293°C/249°C 290°C/249°C 304°C/265°C 304°C/265°C 309°C/265°C 3107266°C 310°C/266°C Steam Temperature 252°C 252°C 256°C 256°C 265°C 260°C 258°C Tube Supports C-steel C-steel C-steel C-steel 410SSS 410SSS 410SSS Lattice Bars Trifoil Trifoil Trifoil Lattice bars Trifoil Trifoil Broached Broached TSP Broached TSP Broached Broached TSP TSP TSP U-bend Supports C-steel (Cu Staggered Stacked C-steel Staggered Flat 41 OS AVBs 410SSS 410SSS bearing) C-steel scallop scallop bars C-steel staggered staggered scallop Lattice bars bars scallop bars scallop bars bars Tube Expansion Hard roll, near secondary face Hydraulic, near Hydraulic Hydraulic secondary face Secondary System Ferrous/Copper Ferrous/Copper Ferrous/Copper All ferrous All ferrous All ferrous; full Ferrous/Copper (Cu being (Cu being flow deep bed removed) removed) polishers Feedtrain Heaters LP: Admiralty LP: Admiralty LP: Admiralty LP: stainless LP: stainless LP: stainless LP: stainless Brass Brass Brass steel steel steel steel HP: 90/10 Cu/N HP: 90/10 HP: 90/10 HP: stainless HP: stainless HP: stainless HP: carbon steel Reheater: Cu/Ni Cu/Ni steel sieel steel Reheater" carbon carbon steel Reheater: Reheater: Reheater: Reheater: Reheater: steel carbon steel carbon steel carbon steel carbon steel carbon steel Preheater Integral Integral Separate Separate Integral Integral Integral Condenser Tubes Admiralty Brass Unit 5 Admiralty Brass 304L SS Titanium Titanium Admiralty Brass (304SS outer) Titanium; Unit (304SS outer) (316 outer) 6 to be changed to titanium Units 7/8 Admiralty Brass (304SS outer tube) Secondary AVT/morph/ AVT/morph/ AVT/morph/ AVT/NH3/ AVT/NH3/N2H Phosphate & AVT/Morpholine Chemistry N2H4 N2H4 N2H4 N2H4 4 Morph/N2H4 Boric Acid Condenser Water Lake Ontario Lake Ontario Lake Huron Lake Huron Lake Ontario Sea water (Bay St. Lawrence ofFundy) River Condensate No No No No No Full-time No Polishing Full-flow -9-

Table 2: Summary of degradation mechanisms found in CANDU steam generators to date.

UNIT PROBLEM CAUSE COMMENTS Pickering-A Pitting/wastage in sludge Deep sludge piles. Crevice cleaning carried out in Units I piles Poor chemistry control (for and 2. SG visually clean; residual tube all Pickering units) for some sheet sludge pile. Pitting mainly in time. Unit 1. RJHT increase Primary side fouling? Primary side cleaning of straight legs Divider plate leakage. in unit 1 produced no improvement in RIHT. Pickering-B Pitting/wastage in sludge Sludge piles: heavy deposits; Feedwater chemistry upgrading taking piles and first support- condenser in-leakage. place. Cu in sludge. Chemical plate broaches Secondary/primary side cleaning of Units 5 and 6 resulted in RIHT increases fouling. Poor chemistry no improvement in RIHT but left SGs control (for all Pickering visually clean. units) for some time. Bruce-A IGSCC/IGA in U-bend at Denting of tubes at scallop Feedwater chemistry upgrading taking scallop bars bar intersections and place. Cu and Pb in sludge. Secondary "jacking" of scallop bars. side chemical cleaning carried out in Lead (Pb) contamination in Units 1. 3 and 4. Unit 2 accelerated cracking. Some fatigue/corrosion fatigue involvement. Shallow pitting in sludge Possibly acidic sulphate Sulphuric acid ingress from water pile conditions. treatment plant in 1985. Boiler level oscillations Fouling of upper support Lancing effective in stopping plate. oscillations. Fatigue Excessive vibration. Scallop bar corrosion Unknown. Possibly related May be some flow-assisted corrosion to crevice corrosion under also. deposits/acidic conditions. Bruce-B Fretting at U-bend and Excessive clearances of U- top support plate bend supports. Shallow pitting in sludge Possible acid and caustic pile excursion (Water Treatment Plant (WTP) in 1989). RIHT increasing Divider plate leakage. Secondary side visually clean. Unit 6 Primary side fouling? has most significant increase. Point Lepreau U-bend fretting/minor U-bend support design? Secondary side clean carried out. pitting in sludge pile/first Underdeposit corrosion. support area RIHT increasing Divider plate leakage. Recovered ~2°C in RIHT. Floating Primary or secondary side one-piece divider plate installed. fouling? Primary side clean carried out on 60% of tubes. Emergency Water Material loss by flow Replacement with low alloy steel. System Header erosion assisted corrosion. Gentilly-2 RIHT increasing Divider plate leakage. Recovered ~2.5°C in RIHT. Floating Primary or secondary side one-piece divider plate installed. Did fouling? not carry out large scale primary side cleaning. - 10-

Table 3: Summary of life management strategy highlights for Canadian CANDU steam generators. UNIT LIFE MANAGEMENT PLAN HIGHLIGHTS Bruce-A Manage existing degradation mechanisms and maximize SG life: • cleaning to remove deposits on tubes, supports, tubesheets • replace all copper components on secondary side • improve water treatment plant • boric acid • wet lay-up recirculation system • high hydrazine chemistry • refurbishment of steam generator internals to remove stresses and reduce flow-induced vibration • extensive tube inspection/plugging • tube removals for metallographic assessments and fitness for service evaluations • hot soaks implemented • new water treatment plant • reduction of condenser and air in-leakage • optimization of pH/amine control Bruce-B Manage existing fretting. Proactive means to prevent corrosion and RIHT increase: • enhance in-service inspection to determine fretting extent/severity • evaluation of effectiveness of prototype U-bend modifications • preparation of enhanced IS1 for non-fretting degradation • tube removals for metallographic assessments and fitness for service evaluations • monitor deposit buildup • evaluate cleaning options Pickering A Manage existing degradation: • secondary side cleaning • condenser retube of air removal zone • improve condenser in-leakage detection and repair • high hydrazine chemistry • routine blowdown • improve water treatment plant • hot soaks • primary side cleaning (Unit 1) • tube removals for metallographic assessments and fitness for service evaluations Pickering B Manage pitting degradation: • extensive inspection/plugging • tube removals for metallographic assessments and fitness for service evaluations • replace copper in feedtrain (including condenser) • high hydrazine chemistry • improve water treatment plant • tube sleeving development • water lancing and chemical clean of all steam generators • improved condenser and air in-leakage detection and repair Darlington NGS Proactive program of maintenance, inspection and good chemistry control: • enhanced in-service inspection to detect degradation as soon as possible • water lancing of all steam generators every four years • process qualification and conceptual design work for chemical cleaning Point Lepreau Manage existing degradation: NGS • secondary cleaning to control corrosion • improve chemistry control (move to AVT?) • remove primary side deposits • replace divider plate Gentilly-2 Maintain clean steam generator and good chemistry control: • replace divider plate -11 -

Table 4: Summary of R&D activities in the COG steam generator program.

TECHNOLOGY APPLICATIONS CHEMISTRY Crevice chemistry /hideout/hideout return for CANDU SGs Sludge Pile Chemistry Molar ratio control studies for CANDU Boric acid addition to the secondary side Zinc additions to the primary circuit (primary side fouling/outlet feeder flow- assisted corrosion (FAC)) Amine chemistry Electrochemical potential (ECP) correlations CORROSION/FATIGUE IGA/IGSCC of Alloys 400, 600, 800, 690 under CANDU conditions Pitting of CANDU SG tubing Corrosion fatigue/fatigue of CANDU SG tubing, especially in U-bend Corrosion of carbon steel supports Corrosion inhibitors Corrosion monitoring FOULING SG fouling mechanisms and predictions (SLUDGE code development) Corrosion product transport characterization, prediction and assessment (on-line CPT monitor development, control system development, oxide solubility measurements) Applications of CHECWORKS to assess FAC contribution to primary and secondary SG tube fouling Additives for reduction of deposition In-situ sludge characterization (laser Raman, X-ray fluorescence) Sludge chemistry RIHT prediction code Thermal properties of tubes and deposits CLEANING Improvements to chemical cleaning processes for CANDU sludge chemistry On-line and dilute cleaning developments Ultrasonically-assisted chemical cleaning Primary side cleaning technology Corrosion monitoring (for cleaning) technology VIBRATION AND FRETTING Vibration and fretting assessments of all CANDU SG designs (PIPO, VIBIC and H3DMAP codes) Fretting fatigue evaluation Underlying vibration and fretting mechanisms THERMALHYDRAULICS Thermalhydraulic evaluations of CANDU SG designs (THIRST code) Evaluation of effects of supports on thermalhydraulics Downcomer flow monitoring technology INSPECTION AND TOOLING Eddy current probes and ultrasonic probes for CANDU SG tubing degradation Eddy current data handling development Validation of eddy current and ultrasonic response (development of probability of detection curves) Qualification of NDE analysts Specialized tooling for SGs (plug leak detector, tube slitting tool, tube sampling tool, sludge sampling tool, i.d. replication tool) TECHNOLOGY TRANSFER Fitness-for-Service guidelines development and database generation. World-wide steam generator survey (in collaboration with EPRI) Specialized symposia on topics of interest to SG owners - 12-

Ontario Hydro Nuclear Station Incapability due to Steam Generators Forced and Planned Outages

Incapability %ICbF 16 16

14 = 14

12 12

10 10

8 8

— 6 — — 6

4 4 — 2 — MM 2 — 0 0 0 7172 73 74 75 76 77787980 81828384858687888990 9192939495

Year

IWForced OPIanned]

176 Steam Generators, 591,712 tubes in total in service

Figure 1: Ontario Hydro Nuclear station incapability due to steam generator forced and planned outages. -13-

1. Understanding SG Ageing 2. SG Operation The key to effective ageing management: • Materials and material properties Managing ageing mechanisms: • Stresses and operating conditions • Follow operating guidelines Minimize^ • Control of water chemistry, • Ageing mechanisms expected impurity incursions, and • Degradation sites degradation deposits • Conditions indicators • Removal of secondary side • Consequences of age-related crevice impurities degradation and failures under normal operating and DBE conditions 1 Check for degradation i 3. SG Inspection, Monitoring, and Assessment

Detecting and assessing ageing effects: • Tubing inspection Feedback • FW nozzle, adjacent piping, shell inspection • Fatigue monitoring • Leak rate monitoring • Fitness-for-service assessment

Correct unacceptable degradation i 4. SG Maintenance: Mitigation, Repair, and Replacement

Managing ageing effects: • Criteria for maintenance work • Mitigation/repair of tube degradation • Vibration control • Mitigation/repair of FW nozzles and adjacent piping • Chemical cleaning • SG replacement

Figure 2: Key elements of a steam generator life management program and associated interfaces. (From IAEA document "Assessment and Management of Ageing of Major Nuclear Power Plant components Important to Safety: Steam Generator", to be published in 1996.) AECL-11912

ISSN 0067-0367 To identify individual documents in the series, we have assigned an AECL- number to each.

Please refer to the AECL- number when requesting additional copies of this document from: Document Centre AECL Chalk River, Ontario Canada K0J1 JO

Fax: (613)584-1745 Tel.: (613) 584-3311 ext. 4623 Price: A

Pour identifier les rapports individuels faisant partie de cette series, nous avons affecte un numero AECL-a chacun d'eux.

Veuillez indiquer le numero AECL- lorsque vous demandez d'autres exemplaires de ce rapport au:

Service de Distribution des Documents Officiels EACL Chalk River (Ontario) CanadaKOJ 1J0

Fax: (613) 584-1745 Tel.: (613) 584-3311 poste 4623 Prix: A